CountFormer: A Transformer Framework for Learning Visual Repetition and Structure in Class-Agnostic Object Counting

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Hossain, Md Tanvir, Islam, Akif, Ameen, Mohd Ruhul
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912953340526592
author Hossain, Md Tanvir
Islam, Akif
Ameen, Mohd Ruhul
author_facet Hossain, Md Tanvir
Islam, Akif
Ameen, Mohd Ruhul
contents Humans can often count unfamiliar objects by observing visual repetition and composition, rather than relying only on object categories. However, many exemplar-free counting models struggle in such situations and may overcount when objects contain symmetric components, repeated substructures, or partial occlusion. We introduce CountFormer, a controlled adaptation of a density-regression framework inspired by CounTR, where the image encoder is replaced with the self-supervised vision foundation model DINOv2. The resulting transformer features are combined with explicit two-dimensional positional embeddings and decoded by a lightweight convolutional network to produce a density map whose integral gives the final count. Our goal is not to propose a new counting architecture, but to study whether foundation-based representations improve structural consistency under a strictly exemplar-free setting. On FSC-147, CountFormer achieves competitive performance under the official benchmark (MAE 19.06, RMSE 118.45). Qualitative analysis suggests fewer part-level overcounting errors for some structurally complex objects, while overall error remains broadly consistent with prior approaches. Sensitivity analysis shows that evaluation metrics are strongly affected by a small number of extreme high-density scenes. Overall, the results highlight the role of representation quality in exemplar-free object counting.
format Preprint
id arxiv_https___arxiv_org_abs_2510_23785
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle CountFormer: A Transformer Framework for Learning Visual Repetition and Structure in Class-Agnostic Object Counting
Hossain, Md Tanvir
Islam, Akif
Ameen, Mohd Ruhul
Computer Vision and Pattern Recognition
Artificial Intelligence
Humans can often count unfamiliar objects by observing visual repetition and composition, rather than relying only on object categories. However, many exemplar-free counting models struggle in such situations and may overcount when objects contain symmetric components, repeated substructures, or partial occlusion. We introduce CountFormer, a controlled adaptation of a density-regression framework inspired by CounTR, where the image encoder is replaced with the self-supervised vision foundation model DINOv2. The resulting transformer features are combined with explicit two-dimensional positional embeddings and decoded by a lightweight convolutional network to produce a density map whose integral gives the final count. Our goal is not to propose a new counting architecture, but to study whether foundation-based representations improve structural consistency under a strictly exemplar-free setting. On FSC-147, CountFormer achieves competitive performance under the official benchmark (MAE 19.06, RMSE 118.45). Qualitative analysis suggests fewer part-level overcounting errors for some structurally complex objects, while overall error remains broadly consistent with prior approaches. Sensitivity analysis shows that evaluation metrics are strongly affected by a small number of extreme high-density scenes. Overall, the results highlight the role of representation quality in exemplar-free object counting.
title CountFormer: A Transformer Framework for Learning Visual Repetition and Structure in Class-Agnostic Object Counting
topic Computer Vision and Pattern Recognition
Artificial Intelligence
url https://arxiv.org/abs/2510.23785